Update faceswap.py

[mst-rajatmishra]

Hi,
I tried Making some improvements and changes,

• Improved error handling for cases where no faces or multiple faces are detected.
• Logs errors in a faceswap.log file for better traceability.
• Replaced the manual image blending with OpenCV’s seamlessClone, which provides more natural blending of the face into the background.
• The code is now more modular with clearly defined functions for specific tasks (e.g., face detection, landmark extraction, transformation, warping, etc.).
• detect_faces handles face detection, and get_landmarks handles landmark extraction, making the flow of the script easier to follow.
• The script uses Python’s logging module to track important information and errors. Logs are written to a faceswap.log file.
• Better user feedback in case of errors, and logging for debugging.
  Thank You,
  RAJAT MISHRA